code hygiene: idiomatic error naming #24
Conversation
| // ByzantineColumnError is thrown when there is a repaired column does not match the expected column merkle root. | ||
| type ByzantineColumnError struct { | ||
| ColumnNumber uint | ||
| LastGoodSquare ExtendedDataSquare |
There was a problem hiding this comment.
Is it really OK to remove this LastGoodSquare? It's currently not really used but I'm not sure if this will be needed to proof that this column was actually byzantine? (cc @adlerjohn) (same for rows below)
In other words, will this be useful to construct these: https://github.com/lazyledger/lazyledger-specs/blob/master/specs/data_structures.md#badencodingfraudproof ?
There was a problem hiding this comment.
I don't really see the need for returning a backup copy of the entire square?
There was a problem hiding this comment.
Yeah, that's what I thought too. Thanks for confirming. The data is in the hands of the caller anyways. If required, the caller can construct everything from the row/col index and the data again (not sure if they need to in any case though).
It would help if the spec actually explained under what circumstances nodes and (which nodes?) would construct a BadEncodingProof from what data. Should I open an issue about that?
There was a problem hiding this comment.
Should I open an issue about that?
Maybe a comment in celestiaorg/celestia-specs#23 that explains what specifically you want to see? I'm going to start working on Evidence relatively soon.
There was a problem hiding this comment.
celestiaorg/celestia-specs#23 (comment)
Is that clear enough?
I've spend some time looking into refactoring and cleaning up rsmt2d. This is the first of a few smaller PRs in this direction:
Err